2 :0 0 p m - 6 :0 0 p m
TUTORIAL V
Neural Network Models for Speech and Image Processing
B.Yegnanarayana
Indian Institute of Technology, Madras

Audience: Any one with an engineering degree, preferably Computer Science or Electrical Engineering, will find this course useful. It is also useful for any practicing engineer or a scien-tist in a research and development establishment and for teachers in academic institutions.

Course Description: Most applications involving speech and images require extraction of infor-mation in the form of features from raw data and use those features for classification, stor-age and retrieval of information. Conventional methods of signal processing use linear meth-ods or some simple nonlinear methods. But in some cases the information is embedded in features which require complex nonlinear pro-cessing of the data for extraction. Moreover, many classification models require nonlinear dividing surfaces in the feature space. Models based on artificial neural networks have been found to be very powerful for feature extrac-tion and classification. This tutorial presents basics of neural network models for feature extraction and classification. In particular, the higher order statistical feature extraction from data, distribution capturing ability, and com-bining evidence from several classifiers, will be discussed in detail. Some applications of these models for processing real speech and image data will be illustrated. In particular, applica-tions for speech enhancement, speech recogni-tion and speaker recognition/verification will be discussed to demonstrate the potential of nonlinear models for these applications. Applications in image processing include image compression, texture analysis and edge extraction, with particular reference to remotely-sensed multispectral data. The course will be self-contained. No specific background of speech and image processing is assumed. The lectures will be illustrated with demon-strations of some speech and vision systems.

Lecturer: B.Yegnanarayana is a Professor at IIT, Madras since 1980. Prior to joining IIT, he was a visiting Associate Professor of Computer Science at Carnegie Mellon University from 1977-1980. He was a member of the faculty at the Indian Institute of Science, Bangalore from 1966 to 1978. He did B. E., M. E., and Ph. D. from IISc, Bangalore, in 1964, 1966, and 1974, respectively. His research interests are in speech, image processing, and neural networks. He has published several papers in these areas in IEEE and other inter-national journals. He is also the author of the book "Artificial Neural Networks", published by Prentice-Hall of India, in 1999. He is a Fellow of the Indian National Academy of Engineering and a Fellow of the Indian National Science Academy.